17387437

Source:http://linkedlifedata.com/resource/pubmed/id/17387437

Download in:

Switch to

Custom View

Named Graph Language Inference

Statements in which the resource exists as a subject.
Predicate	Object
rdf:type	pubmed:Citation
lifeskim:mentions	umls-concept:C0008902, umls-concept:C0023963, umls-concept:C0086022, umls-concept:C0183683, umls-concept:C0220806, umls-concept:C0344211, umls-concept:C0442335, umls-concept:C1171411, umls-concept:C1317973, umls-concept:C1521721, umls-concept:C1705099
pubmed:issue	5
pubmed:dateCreated	2007-4-20
pubmed:abstractText	We investigate the classification performance of circular fingerprints in combination with the Naive Bayes Classifier (MP2D), Inductive Logic Programming (ILP) and Support Vector Inductive Logic Programming (SVILP) on a standard molecular benchmark dataset comprising 11 activity classes and about 102,000 structures. The Naive Bayes Classifier treats features independently while ILP combines structural fragments, and then creates new features with higher predictive power. SVILP is a very recently presented method which adds a support vector machine after common ILP procedures. The performance of the methods is evaluated via a number of statistical measures, namely recall, specificity, precision, F-measure, Matthews Correlation Coefficient, area under the Receiver Operating Characteristic (ROC) curve and enrichment factor (EF). According to the F-measure, which takes both recall and precision into account, SVILP is for seven out of the 11 classes the superior method. The results show that the Bayes Classifier gives the best recall performance for eight of the 11 targets, but has a much lower precision, specificity and F-measure. The SVILP model on the other hand has the highest recall for only three of the 11 classes, but generally far superior specificity and precision. To evaluate the statistical significance of the SVILP superiority, we employ McNemar's test which shows that SVILP performs significantly (p < 5%) better than both other methods for six out of 11 activity classes, while being superior with less significance for three of the remaining classes. While previously the Bayes Classifier was shown to perform very well in molecular classification studies, these results suggest that SVILP is able to extract additional knowledge from the data, thus improving classification results further.
pubmed:language	eng
pubmed:journal	http://linkedlifedata.com/resource/pubmed/journal/8710425
pubmed:citationSubset	IM
pubmed:chemical	http://linkedlifedata.com/resource/pubmed/chemical/Pharmaceutical Preparations
pubmed:status	MEDLINE
pubmed:month	May
pubmed:issn	0920-654X
pubmed:author	pubmed-author:AminiAtaA, pubmed-author:BenderAndreasA, pubmed-author:CannonEdward OEO, pubmed-author:GlenRobert CRC, pubmed-author:MitchellJohn B OJB, pubmed-author:MuggletonStephen HSH, pubmed-author:SternbergMichael J EMJ
pubmed:issnType	Print
pubmed:volume	21
pubmed:owner	NLM
pubmed:authorsComplete	Y
pubmed:pagination	269-80
pubmed:meshHeading	pubmed-meshheading:17387437-Bayes Theorem, pubmed-meshheading:17387437-Computational Biology, pubmed-meshheading:17387437-Confidence Intervals, pubmed-meshheading:17387437-Drug Design, pubmed-meshheading:17387437-Pharmaceutical Preparations, pubmed-meshheading:17387437-Software
pubmed:year	2007
pubmed:articleTitle	Support vector inductive logic programming outperforms the naive Bayes classifier and inductive logic programming for the classification of bioactive chemical compounds.
pubmed:affiliation	Unilever Centre for Molecular Science Informatics, Department of Chemistry, University of Cambridge, Lensfield Road, Cambridge, UK.
pubmed:publicationType	Journal Article, Comparative Study, Research Support, Non-U.S. Gov't